Going beyond the Django ORM limitations with Postgres by Craig Kerstiens

Transcript

1. G!"#$ b%&!#' ()% D*+#$! ORM ,"-"(+("!#. w"() P/($r%. @0r+"$1%r.("%#.
2. Postgres.app PSA: Macs
3. Why Postgres http://www.craigkerstiens.com/2012/04/30/why-postgres/ https://speakerdeck.com/craigkerstiens/postgres-demystified “Its the emacs of databases”
4. Datatypes Conditional Indexes Transactional DDL Foreign Data Wrappers Concurrent Index Creation Extensions Common Table Expressions Fast Column Addition Listen/Notify Table Inheritance Per Transaction sync replication Window functions NoSQL inside SQL Momentum TLDR
5. Limitations? Django attempts to support as many features as possible on all database backends. However, not all database backends are alike, and we’ve had to make design decisions on which features to support and which assumptions we can make safely.
6. D+(+(2%.
7. __________ < Postgres > ---------- \ ^__^ \ (oo)\_______ (__)\ )\/\ ||----w | || ||
8. Datatypes smallint bigint integer numeric float serial money char varchar text bytea timestamp timestamptz date time timetz interval boolean enum point line polygon box circle path inet cidr macaddr tsvector tsquery array XML UUID
9. Datatypes smallint bigint integer numeric float serial money char varchar text bytea timestamp timestamptz date time timetz interval boolean enum point line polygon box circle path cidr macaddr tsvector tsquery array XML UUID inet
10. Datatypes smallint bigint integer numeric float serial money char varchar text bytea timestamp timestamptz date time timetz interval boolean enum point line polygon box circle path cidr macaddr tsvector tsquery array XML UUID inet
11. Datatypes smallint bigint integer numeric float serial money char varchar text bytea timestamp timestamptz date time timetz interval boolean enum point line polygon box circle path cidr macaddr tsvector tsquery array XML UUID inet
12. _______ < SQLite > --------- \ ^__^ \ (oo)\_______ (__)\ )\/\ ||----w | || ||
13. Datatypes null real text blob integer
14. Datatypes smallint bigint integer numeric float serial money char varchar text bytea timestamp timestamptz date time timetz interval boolean enum point line polygon box circle path inet cidr macaddr tsvector tsquery array XML UUID
15. I#'3%.
16. ________ < Postgres > ---------- \ ^__^ \ (oo)\_______ (__)\ )\/\ ||----w | || ||
17. Indexes Multiple Types B-Tree, Gin, Gist, KNN, SP-Gist CREATE INDEX CONCURRENTLY
18. _____ < MySQL > ------- \ ^__^ \ (oo)\_______ (__)\ )\/\ ||----w | || ||
19. Indexes They exist
20. Digging In
21. Arrays CREATE TABLE item ( id serial NOT NULL, name varchar (255), tags varchar(255) [], created_at timestamp );
22. Arrays CREATE TABLE item ( id serial NOT NULL, name varchar (255), tags varchar(255) [], created_at timestamp );
23. Arrays INSERT INTO item VALUES (1, 'Django Pony', '{“Programming”,”Animal”}', now()); INSERT INTO item VALUES (2, 'Ruby Gem', '{“Programming”,”Jewelry”}', now());
24. Arrays INSERT INTO item VALUES (1, 'Django Pony', '{“Programming”,”Animal”}', now()); INSERT INTO item VALUES (2, 'Ruby Gem', '{“Programming”,”Jewelry”}', now());
25. Django pip install djorm-ext-pgarray pip install djorm-ext-expressions models.py: from djorm_pgarray.fields import ArrayField from djorm_expressions.models import ExpressionManager class Item(models.Model): name = models.CharField(max_length=255) tags = ArrayField(dbtype="varchar(255)") created_at = models.DateTimeField(auto_now=True)
26. Django pip install djorm-ext-pgarray pip install djorm-ext-expressions models.py: from djorm_pgarray.fields import ArrayField from djorm_expressions.models import ExpressionManager class Item(models.Model): name = models.CharField(max_length=255) tags = ArrayField(dbtype="varchar(255)") created_at = models.DateTimeField(auto_now=True)
27. Django i = Item( name='Django Pony', tags=['Programming','Animal']) i.save() qs = Item.objects.where( SqlExpression("tags", "@>", ['programming']) )
28. Django i = Item( name='Django Pony', tags=['Programming','Animal']) i.save() qs = Item.objects.where( SqlExpression("tags", "@>", ['programming']) )
29. dblink hstore citext ltree isn cube pgcrypto tablefunc uuid-ossp earthdistance trigram fuzzystrmatch pgrowlocks pgstattuple btree_gist dict_int dict_xsyn unaccent Extensions
30. NoSQL in your SQL CREATE EXTENSION hstore; CREATE TABLE item ( id integer NOT NULL, name varchar(255), data hstore, );
31. NoSQL in your SQL CREATE EXTENSION hstore; CREATE TABLE item ( id integer NOT NULL, name varchar(255), data hstore, );
32. NoSQL in your SQL CREATE EXTENSION hstore; CREATE TABLE item ( id integer NOT NULL, name varchar(255), data hstore, );
33. NoSQL in your SQL INSERT INTO items VALUES ( 1, 'Pony', 'rating => "4.0", color => “Pink”', );
34. NoSQL in your SQL INSERT INTO items VALUES ( 1, 'Pony', 'rating => "4.0", color => “Pink”', );
35. Django pip install django-hstore from django.db import models from django_hstore import hstore class Item(models.Model): name = models.CharField(max_length=250) data = hstore.DictionaryField(db_index=True) objects = hstore.Manager() def __unicode__(self): return self.name
36. Django pip install django-hstore from django.db import models from django_hstore import hstore class Item(models.Model): name = models.CharField(max_length=250) data = hstore.DictionaryField(db_index=True) objects = hstore.Manager() def __unicode__(self): return self.name
37. Django Item.objects.create( name='Django Pony', data={'rating': '5'}) Item.objects.create( name='Pony', data={'color': 'pink', 'rating': '4'})
38. Django Item.objects.create( name='Django Pony', data={'rating': '5'}) Item.objects.create( name='Pony', data={'color': 'pink', 'rating': '4'})
39. Django colored_ponies = Product.objects.filter(data__contains='color') print colored_ponies[0].data['color'] favorite_ponies = Product.objects.filter(data__contains={'rating': '5'}) print colored_ponies[0]
40. Django colored_ponies = Product.objects.filter(data__contains='color') print colored_ponies[0].data['color'] favorite_ponies = Product.objects.filter(data__contains={'rating': '5'}) print colored_ponies[0]
41. Q4%4%"#$
43. Pub/Sub?
44. Postgres a great Queue
45. Postgres a great Queue Not With Polling
46. Trunk pip install celery trunk psql < sql/*.sql celery worker -A tasks --loglevel=info ipython -i tasks.py >>> add.delay(2, 2)
47. Trunk pip install celery trunk psql < sql/*.sql celery worker -A tasks --loglevel=info ipython -i tasks.py >>> add.delay(2, 2)
48. Trunk from celery import Celery celery = Celery('tasks') celery.config_from_object({ 'BROKER_URL': 'trunk.transport.Transport:// localhost/trunk', }) @celery.task def add(x, y): return x + y
49. Trunk from celery import Celery celery = Celery('tasks') celery.config_from_object({ 'BROKER_URL': 'trunk.transport.Transport:// localhost/trunk', }) @celery.task def add(x, y): return x + y
50. T3( S%+r0)
51. Searching Text
52. Searching Text Lucene Sphinx Elastic Search Solr
53. Searching Text Lucene Sphinx Elastic Search Solr Postgres
54. Full Text Search CREATE TABLE posts ( id serial, title varchar(255), content text, tags varchar(255)[], post_text tsvector ); CREATE INDEX posttext_gin ON posts USING GIN(post_text); CREATE TRIGGER update_posttext BEFORE INSERT OR UPDATE ON posts FOR EACH ROW EXECUTE PROCEDURE tsvector_update_trigger( ‘PostText’,‘english’,title, content, tags);
55. Full Text Search CREATE TABLE posts ( id serial, title varchar(255), content text, tags varchar(255)[], post_text tsvector ); CREATE INDEX posttext_gin ON posts USING GIN(post_text); CREATE TRIGGER update_posttext BEFORE INSERT OR UPDATE ON posts FOR EACH ROW EXECUTE PROCEDURE tsvector_update_trigger( ‘PostText’,‘english’,title, content, tags);
56. Django from djorm_pgfulltext.models import SearchManager from djorm_pgfulltext.fields import VectorField from django.db import models class Page(models.Model): name = models.CharField(max_length=200) description = models.TextField() search_index = VectorField() objects = SearchManager( fields = ('name', 'description'), config = 'pg_catalog.english', search_field = 'search_index', auto_update_search_field = True )
57. Django from djorm_pgfulltext.models import SearchManager from djorm_pgfulltext.fields import VectorField from django.db import models class Page(models.Model): name = models.CharField(max_length=200) description = models.TextField() search_index = VectorField() objects = SearchManager( fields = ('name', 'description'), config = 'pg_catalog.english', search_field = 'search_index', auto_update_search_field = True )
58. Django Page.objects.search("documentation & about") Page.objects.search("about | documentation", raw=True)
59. Django Page.objects.search("documentation & about") Page.objects.search("about | documentation", raw=True)
60. I#'3%.
61. Indexes B-Tree Generalized Inverted Index (GIN) Generalized Search Tree (GIST) K Nearest Neighbors (KNN) Space Partitioned GIST (SP-GIST)
62. Indexes Which do I use?
63. Indexes B-Tree Generalized Inverted Index (GIN) Generalized Search Tree (GIST) K Nearest Neighbors (KNN) Space Partitioned GIST (SP-GIST)
64. Generalized Inverted Index (GIN) Use with multiple values in 1 column Array/hStore
65. Generalized Search Tree (GIST) Full text search Shapes PostGIS
66. Indexes B-Tree Generalized Inverted Index (GIN) Generalized Search Tree (GIST) K Nearest Neighbors (KNN) Space Partitioned GIST (SP-GIST)
67. G%!Sp+("+,
68. GeoDjango
69. O#% M!r%
70. Connections
71. Connections django-postgrespool djorm-ext-pool django-db-pool
72. django-postgrespool import dj_database_url import django_postgrespool DATABASE = { 'default': dj_database_url.config() } DATABASES['default']['ENGINE'] = 'django_postgrespool' SOUTH_DATABASE_ADAPTERS = { 'default': 'south.db.postgresql_psycopg2' }
73. django-postgrespool import dj_database_url import django_postgrespool DATABASE = { 'default': dj_database_url.config() } DATABASES['default']['ENGINE'] = 'django_postgrespool' SOUTH_DATABASE_ADAPTERS = { 'default': 'south.db.postgresql_psycopg2' }
74. Limitations? Django attempts to support as many features as possible on all database backends. However, not all database backends are alike, and we’ve had to make design decisions on which features to support and which assumptions we can make safely.
75. D*+#$! ORM Its not so bad
76. P/($r%. Its great
77. 5+#1.!